Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    conversational

    np_max-act · gemini-2.0-flash

    The neuron is primarily sensitive to interrogative phrases, especially the “What is it with …?” question pattern.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    iset
    -0.07
     convers
    -0.06
     cautious
    -0.06
    /fa
    -0.06
    .big
    -0.06
    iene
    -0.06
     accelerate
    -0.06
     Ere
    -0.06
     respondent
    -0.06
     alpha
    -0.06
    POSITIVE LOGITS
     Connections
    0.07
     #(
    0.06
     평
    0.06
    ODY
    0.06
    قر
    0.06
    RoutingModule
    0.06
     استرات
    0.06
    @Before
    0.06
     Ranked
    0.06
     ROS
    0.06
    Act Density 0.149%

    No Known Activations